The kinetics of powder consolidation, or densification, and the powder morphological changes ocurring during hot isostatic pressing (HIP) are studied as a function of particle size distribution and hold time at HIP temperature for the nickel base superalloy RENE-95. In order to understand the extent of individual powder particle deformation during consolidation and its effect on subsequent prior particle boundaries (PPB), particle size distribution was studied as a variable. Particle size distributions studied include monosized (75–90 um), bimodal ( 75–90 um and 33–35 um) and commercial (<104 um) size distributions. The experimental results of HIP densification kinetics are compared with a newly developed analytical deformation mechanism model for HIP consolidaiton which takes into account the effect of a distribution of particle sizes on the kinetics of densification.

Powders of Al alloy 7091 have been consolidated by means of dynamic compaction. The dependence of density and hardness on projectile velocity has been determined. The resulting as-compacted material has been characterized using analytical transmission electron microscopy and evidence of interparticle melting observed. The microstructural responseof the compacted material to heat treatment at 523 and 723°K has been investigated.

Hardness and ductility in Fe-M-B alloys were found to be directly dependent on the strength of the metal-metalloid bonding. This behaviour, as well as the annealing embrittlement, can be explained by the localized viscous flow model of heterogeneous deformation.

Solidification of metal alloys by rapid quenching can result in the formation of amorphous or microcrystalline solids, or materials with improved microstructural homogeneity, all with the view of forming new phases or obtaining improved properties. Some alloys may be cooled at high rates to achieve “microcrystallinity” but cannot be cooled rapidly enough to become amorphous. Other alloys may achieve both conditions depending on the cooling rate. We have examined the effects of cooling rate on the structure of one alloy that can, depending on the cooling rate, be made partially or completely amorphous. The alloy is Fe75Si15B10 (atom percent) which was formed as ribbon by melt spinning and as powder by spark erosion in dielectrics of different cooling characteristics and by gas-water atomization. The structural characteristics were determined by x-ray diffraction, measurements of magnetic properties and by optical and scanning electron microscopy.

The influence of short range order, electron concentration and size misfit on the thermal stability of (Fe100-xMx)83B17 metallic glasses has been investigated (M = Cr, Mn, Co, Ni, Mo, W). The results are discussed together with Curie-temperature measurements and Mößbauer spectroscopy data. The concentration range in which the various stable and metastable phases form is given.

The melt spinning technique was used to prepare composite materials consisting of a metallic glass matrix and a second phase in the form of particles and/or C or SiC fibers. The spatial distribution of the second phase varied with the method used to incorporate the particles into the melt. The observed distributions can be explained by assuming that rotational stirring takes part in the melt puddle. The influence of second phase on the effective quenching rate is discussed and the mechanical properties of Ni75Si8B17 glass containing a variety of second phase particles are reported.

It is proposed that the effects of crystalline inclusions on the mechanical properties of metallic glasses are due to the presence of compressive stress-fields arising from the difference in thermal expansion coefficients of the crystals and the glass. Because of the dilatational micro-mechanism of plastic flow in metallic glasses these compressive stresses raise the tensile fracture stress but reduce the hardness. The temperature dependence of the hardness changes in partially crystallised Fe40Ni40P14B6 is consistent with this model.

Sub-critical crack growth is a function of the degree of relaxation of the amorphous structure and has implications regarding the measurement of annealing embrittlement and the dependence of embrittlement on temperature. The variation of Kc with annealing time is strongly dependent on the size of the initial crack length. Quenched-in inhomogeneities seem to be always present. Their effect depends on the specific testing conditions, especially the testing temperature because the ductile-brittle transition is a function of the degree of relaxation of the amorphous structure. Some doubt is also cast on the accepted assumption that a chevron pattern is representative of plane strain crack propagation conditions.

The corrosion behaviour in liquid electrolytes were investigated for some Fe- and Ni-based metallic glasses varying pH-value, electrolyte temperature and heat treatment. For some (Fe100-xMx)83B17 glasses the high temperature oxidation behaviour was studied.

Rapidly solidified ribbons of Fe67Ti13P(20-x)Bx (x= 0, 7, 20 a/o) and Fe(85-y)Ti l5Py (y=5, 10 a/o) alloys were produced by melt spinning, and their structure and anodic polarization behavior were examined by x-ray diffraction and potentio-dynamic polarization measurements in deaerated 1 N H2SO4, respectively. The only alloy that developed an amorphous structure on rapid solidification, Fe67Ti13B20, exhibited poor corrosion resistance, whereas crystalline alloys with 10 a/o or more P exhibited a tendency to passivate during anodic polarization. Substituting Cr for Ti led to significant improvement in corrosion resistance and promoted development of glassy structures in Fe67(TiCr)13P13B7 compositions. No such improvements were observed when Cu was substituted for Ti. It is concluded that alloy composition is more important than degree of crystallinity in determining corrosion resistance.

Phosphorous has been shown to have a beneficial effect on reducing the corrosion rate of various metal alloys. This research was undertaken to determine by what mechanism the phosphorous is able to improve the corrosion resistance of metals. The amorphous copper-zirconium system was chosen for experimentation because the amorphous structure eliminates the complications due to crystalline defects. The zirconium forms zirconia (ZrO2), which is normally a protective surface oxide. Copper-zirconium alloys with and without phosphorous were melt-spun, and were tested in various electrolytes. In comparing Cu-42Zr-2P with Cu-4OZr, the results show that the corrosion rate of Cu-42Zr-2P is lower in 1N sulfuric acid, and significantly lower in 3.5 w/o NaCl, and in 1N HCl. ESCA analysis shows that phosphorous is present in the zirconia film, and has a valence of +5. Photocurrent experiments suggest that phosphorous alters the semiconductor properties of the zirconia film.

A series of rapidly solidified Ni-Mo-Cr-B alloys have been investigated for possible application as corrosion resistant hardfacing coatings. As-cast amorphous ribbons were pulverized to powder, which was subsequently applied to a substrate by plasma spraying. The coatings are substantially amorphous, as revealed by x-ray diffraction and transmission electron microscopy. They exhibit ∼95% of theoretical density and exhibit good corrosion and wear resistances, superior to those of many commercially available hardfacing coatings.

High-precision density measurements on coldrolled Pd77.5Cu6Si16.5 glass rods before and after relaxation anneals showed that 30 – 40% rolling leads to a density reduction of ∼0.14%, recoverable on subsequent annealing below Tg. This change is about half of the density increase attainable by recovery-annealing the as-quenched glass.

Density changes in Fe40Ni40B20 glass ribbon, resulting from recovery-annealing, were studied indirectly by measuring isothermal length changes through dilatometry of foil sandwiches. The experimental conditions for reliable isothermal dilatometry, using a commercial instrument, are outlined. It was found that the kinetics of length change in the longitudinal and transverse directions were quite different, being faster in the former. Prolonged recovery-annealing prior to dilatometry removed this difference. The intermediate stage of each isotherm followed loqt kinetics.

Experiments were performed with Fe40Ni40B20, Ni64Zr36 and Cu66Ti34 to search for a reversible component of length change during cyclic isothermal holds at two temperatures. Such a component probably exists but is extremely small, less than 1% of the limiting irreversible length change.

The implications of the various findings are discussed in the light of recent literature.

Changes in physical and mechanical properties were correlated with relaxation phenomena. A quantitative description of the process is possible. Due to the anisotropy, relaxation in a particular direction does not give the same results as volume properties.

Low temperature relaxation of Fe40Ni40B20 glassy samples was monitored by DSC, velocity of sound and creep recovery measurements. The as-quenched samples exhibited very pronounced exothermic reactions in the temperature region prior to crystallization. Reversible changes were found during subsequent heat treatmentof preannealed glassy material having exothermic and endothermic heat flow during ordering and disordering respectively. Results were found to be in agreement with velocity of sound and creep recovery changes, found earlier , having activation energy for these reversible changes around 160 kJ/mole.

Pd-Y-Si glasses with a Si content as low as 9 at.% were produced by hammer and anvil quenching, the reduction of the Si concentration from a minimum value of 14 at.% in binary Pd-Si alloys was due to the synergistic effect of the addition of Y. The glass forming composition region is reported.

The alloy Pd2Y,Si, was studied in detail. The change of the microhardness upon isothermal annealing was investigated. Isothermal and non-isothermal heating experiments yielded activation energies of 52 and 53 kcal/mol (218 and 222 kJ/mol) in good agreement with each other and an Avrami exponent n = 3.7 was found.

The short melt duration resulting from pulsed laser and electron beam surface melting of ion-implanted metals has been used to measure precipitate nucleation times of compounds within the melt. We have examined the phases present in several alloy systems with TEM and used calculated thermal histories to place limits on the time required for nucleation of the following compounds: AlSb (5–25 ns), Al3Ni (≳ 750 ns)Al3Ni2 (≳ 950 ns) and AlNi (< 1000 ns), all in molten Al, and TiC (≲ 100 ns) in molten Fe. The compounds observed after our rapid solidification have relatively simple, cubic structures and melt congruently, while those predicted but not observed have more complex structures and decompose peritectically.

A metallic glass ribbon of Cu55Ti45 prepared by melt spinning was examined by x-ray, neutron, and electron diffraction, by small angle neutron diffraction (SANS), transmission electron microscopy (TEM), and by differential thermal analysis (DTA). In the liquid quenched condition large angle diffraction data (both x-ray and neutron) show the broad banded structure typical of the amorphous state. The SANS data, however, exhibit highly anisotropic patterns arising from the phase decomposition during solidification. Ribbons annealed below the glass transition temperature (Tg ) produced neutron diffraction patterns of materials with the same amorphous structure combined with a new short range order; and the SANS patterns retained the asymmetry of the as-quenched material. Ribbons annealed above the crystallization temperature (Tc) show both isotropic and anisotropic contributions to the SANS patterns. Formation of the equilibrium TiCu phase occurs directly from the metallic glass at Tc. The equilibrium Ti3Cu4 phase, however, forms from the TiCu phase at slightly higher temperatures.

Recently, significant advances have been made in establishing a theoretical basis for explaining the many structural-sensitive physical properties of amorphous materials. Viewed from both the theoretical and experimental realms, reversible annealing dependency and the structural relaxation mechanism of the amorphous solids are conceptually deducible from the concept of localized structural fluctuations. In the case of annealing, however, the limiting temperature below which no permanent atomiclevel rearrangements occur is not known and may well be dependent on several intrinsic properties of the amorphous state. In an effort to establish this limiting temperature, amorphous Cu60-Zr 40 samples have been annealed using a 1kv pulsed laser. Samples annealed during 1200, 800, 600 400, and 200 μ sec. have been analyzed by using Differential Scanning Calorimetry (DSC), High Resolution Electron Microscopy (HREM), and microanalysis. Analysis of the DSC indicates variations in both height and the shape of the exotherms from the samples annealed above 600 μ sec. duration. The results from HREM of these samples indicate the onset of amorphous-crystalline transition evidenced during 200 μ sec. annealing.

An attempt has been made to influence the trend of crystallization of a solidifying melt through the crystallization of an amorphous substrate undergoing transformation in its contact at the solid liquid interface. Two cases have been studied. In one case involving amorphous Cu60-Zr 40/Sn interface, it has been found that both textural and microstructural characteristics of the substrate, at the onset of the amorphous-crystalline transition, exert profound influence in crystallization, of the solidifying melt. High resolution electron microscopy and diffraction reveal that the preferred textural orienzation of the solidified tin is (111) at the interface and (100) far from the interface. In addition, an abundant array of edge dislocations found at the interface points to the role that these defects have played in the nucleation process of the solidifying melt. In the case of Ni60 Nb40/A1 the textural characteristics of the solidifying aluminum melt appear to be less perturbed by the crystallization effects of the substrate.